Yes, that's very much the way I'm thinking about the question. It all has to do with the fact that we can standardize the measurement of certain things-- and we are then limited to try and cobble together a theory of everything by connecting these measurables. To me, that's kind of like trying to explain why a painting is great by looking at the pigments it used and the ratios of various proportions in the figures depicted. There may be much to be learned about the art of painting by considering such measurables, but they will never serve to reconstitute that art in its entirety.

To sum that up, one could define science as the study of the projection of reality onto the objectively measurable domain, but there is little or no reason to expect that projection to produce a one-to-one restoration of all that is happening "in reality". There is, however, the legitimate question of what elements of reality survive that projection, and what elements do not. I expect that a true "theory of everything" would not survive such a projection, but a more apropos question is, can there even be a theory of all the things that do survive the projection, or do the most sublime elements underlying even the projection require something that does not survive it? That's the element that I think people like Dawkins are overlooking.