Steve: As you know Lewis opposed reductionistic accounts of even physical phenomena. The "physical thing" we call the sunbeam has nonphysical properties which have to be siphoned off in order to make it a physicalistic account. The concept of the physical is supposed to be a) mechanistic, and b) closed and c) everything else has to supervene on that. At least that's the Hasker-Reppert definition of physicalism, which can be expanded to come up with an account of what naturalism is supposed to be. (We are happy to solve the physicalists' problem of defining themselves for them, and I have it on the authority of Blue Devil Knight that our definition is a good one).
If you look at even secular philosophers like Nagel who take the toolshed distinction seriously, we find that they push the limits of what is acceptable as physicalism. Nagel seems to have broken out of physicalism even if he hasn't found his way to theism (neither did Lewis when he accepted the argument), and while Searle tries to be a materialist, I think most people on both sides of the materialist debate think he fails to do so. The strongest physicalists like Dennett, Churchland, and company, including ordinary functionalists like the early Putnam try their best to explain the distinction away.
Materialism is an attempt to say that the world as analyzed by the senses and the method of science is the ultimate reality, and that everything else is a byproduct. Insofar as they are consistent materialists, they have to try to undercut Lewis's looking at-looking along distinction. you can't press it into the service of physicalism without begging the question.