In the philosophical spirit of qualifying everything to death, I want to clarify the limits I see to the metaphor of the manifest and scientific images of humankind, lest it seem I naively endorse every aspect of this metaphor. With these qualifications in mind, I then want to sermonize a bit about why I think the metaphor is still important.
Sellars himself speaks of this metaphor in several conflicting ways. For example, he compares the synoptic, stereoptic perspective of the two images to two perspectives from which one can see a landscape as a whole. But while this alternative metaphor emphasizes the kind of metaphorical depth Sellars is seeking to gain from philosophical inquiry, it obscures philosophy's distinctive subject matter, since landscapes, by definition, contain no human subjects. In landscapes, different perspectives just clarify a unitary, three-dimensional impersonal scene, but the whole point of Sellars' two images is that the images are rival images of humankind and so cannot be easily accommodated into a single coordinate system without a lot of cropping and adjustment.
Alternatively, Sellars explains that his metaphor of images is an oblique way of talking about "conceptions" of humankind: "for I am using 'image' in this sense as a metaphor for conception." And strangely enough, some critics have accused Sellars' metaphor of images as being "inept" because he is really talking about concepts (e.g. in Life and Action [2008: 10], Michael Thompson intimates that Sellars has made a mistake: "The expression 'image' is inept; Sellars is contrasting certain systems of concepts."). But Sellars surely realized that talking about systems of concepts will not make the point of his essay, that is, will not easily express the goal of philosophy because it loses the metaphor of synoptic or stereoptic vision. How can one "see" two rival systems of concepts in a "stereoptic" view? Mixing these metaphors just obscures his point.
So, Sellars' metaphor clearly has limits: on the one hand, Sellars is really talking about concepts, but it's difficult to articulate the point of philosophical inquiry with such language; on the other hand, Sellars uses a visual metaphor to articulate the point of philosophical inquiry, but this visual metaphor doesn't quite capture what is distinctive about philosophical subject matter, namely, its interest in humankind and self-understanding ("know thyself," as Socrates said). Therefore, like most metaphors, one shouldn't take Sellars' metaphor too seriously, or try to draw too much from it. It provides a rough and ready understanding of what philosophical inquiry is all about.
But despite its fuzziness, Sellars' metaphor strikes me as capturing an important insight, too often missed by otherwise very capable philosophers. For me, one source of constant frustration in reading philosophers like Jerry Fodor, Jason Stanley, Tim Williamson, and others -- who are all far more capable philosophers than I -- is that they seem to lose sight of the goal of philosophy as Sellars articulates it in his essay:
"the aim of philosophy, abstractly formulated, is to understand how things in the broadest possible sense of the term hang together in the broadest possible sense of the term. ... To achieve success in philosophy would be, to use a contemporary turn of phrase, to 'know one's way around' with respect to all these things, not in that unreflective way in which the centipede of the story knew its way around before it faced the question, 'how do I walk?', but in that reflective way which means that no intellectual holds are barred."
Perhaps these philosophers disagree about the goal of philosophy; perhaps they think that philosophy isn't about understanding or know how; it's a much more intellectual matter of knowing that the facts are as they are. However that may be, it seems to me (I'm sure that at some level this is unfair to these extremely thoughtful philosophers, from whom I have learned a great deal), they aren't terribly interested in this kind of getting about in the world "no intellectual holds barred." They seem to me to throw up a lot of roadblocks to more complete self-understanding at various points -- "well, that's just how science does it" or "yes, yes, yes, this metaphysical problem has suspect foundations, but which metaphysical solution is the true one?" While they might never say anything like this outright (well, maybe except Fodor), I sense a lot of impatience in their writing when it comes to diagnosing philosophical problems because they want to get back to what they think is the "real" philosophical question about who is right and who is wrong about the original philosophical problem, never mind whether that philosophical problem is worth answering in the first place.
The example of the foundations of modal discourse illustrates this lack of interest nicely, I think. Philosophers in the late twentieth century have, I submit, become uncritically accepting of modal discourse. Throughout the twentieth century, especially in the latter half, modal logic was formalized in an extremely perspicuous fashion that revealed insights into modal relationships and enabled more expressive modal discourse. As I understand it, Carnap is largely responsible getting the interest going in the early part of the twentieth century as a response to conflicts among Empiricists in the Vienna Circle about the status of modal concepts like mass, charge, and radioactivity in science. Kripke is responsible for generating intense interest in the later half by developing a semantics for Carnap's and others' earlier syntactic formalizations: Kripke proposed that possibility and necessity be treated as quantifiers over "possible worlds" and that what-here-to-fore had been considered implication relations among propositions be treated as "accessibility relations" between such possible worlds. Because modal logic only has one inference rule, Kripke's semantics enabled a much better understanding of the relationships between various axiomitizations of modal logic. But it also generated new problems: what are these new entities, possible worlds? And thus new schools sprang up, Actualists and Possibilists, with competing answers to this question. Philosophers then set about trying to figure out which school was right, but in the process, they gave up their original interest in whether modal discourse was even justified -- they just began assuming it was and went about trying to investigate its metaphysical credentials.
As Brandom has forcefully argued in Between Saying and Doing (a work basically commissioned by Williamson), Kripke's semantics just kicked the original problem about the status of modal discourse in science down the street a bit: the new entities that Kripke postulated, possible worlds, were just as mysterious as the original modal properties they were supposed to explain; and our epistemic access to these new entities was even more mysterious. Given this fact, it's hard to understand how philosophers shifted from a skeptical attitude about modal discourse to an uncritically accepting attitude in one or two generations. To say "well science presumes modal properties, and we'll just take whatever scientific explanation works and then try to investigate it further" (which is frequently the spirit, if not the letter, of Fodor's writings), is to misunderstand the philosophical problem that was faced and to miss an opportunity for philosophical diagnosis and self-understanding characteristic of "no intellectual holds barred" aspirations that mark out philosophy as a distinctive enterprise. In other words, when we are constantly turning to arguments about who-is-right-about-X, and never, "why do we even care about X?" we aren't really doing philosophy as such; we're doing something else. Given this fact, it's not surprising that a lot of philosophers in the last twenty years have begun devoting more time and attention to metaphilosophy (i.e. what is philosophy all about?). This, I think, is a question worth asking, precisely because it is the sort of no-intellectual-holds-barred question that philosophy is in the business of asking in the first place.
(NB: Apologies for these superficial reactions to Fodor, Stanley, and Williamson and for simplifying their views, probably unfairly, but inadvertently so.)