[String phenomenology isn't really my field, but I certainly haven't had a lot of luck deriving a unique background from pure theory. I did once think that I'd found a model that predicted the cosmological constant to within an order of magnitude or two, but that obviously didn't pan out.]
[1] https://en.wikipedia.org/wiki/Theory_of_everything#Infinite_...
Another possible resolution is a cyclic picture (something like https://en.wikipedia.org/wiki/Cyclic_model), in which the cause of the cause of the cause of … may be the original event.
I find myself leaning towards causation breaking down at some point. Just like our concepts of "position" and "momentum" break down at the quantum level.
I can certainly see the sense in which it might be said to be, but I think that's not the sense in which the term is originally meant. I tried to come up with an explanation of why I think they were meaningfully different by an appeal to mathematical analogy, but I'm not sure that it's convincing.
Fermat argued as follows: a positive integer solution to `x^4 + y^4 = z^4` could always be converted into a smaller such solution, and you can't produce ever-smaller positive integers forever, so there weren't any such solutions in the first place. I'd say that this is clearly an infinite regress (though it was actually called infinite descent).
On the other hand, consider the problem of subtracting 1. In the system of ordinary (not-necessarily-positive) integers, this leads to a (non-paradoxical) infinite regress; whereas, in the system of integers modulo some fixed positive integer, it leads to a (non-paradoxical) circularity (where repeatedly subtracting 1 from a number will eventually yield that number itself). I think that it is reasonable to call these different phenomena.
There are interesting questions in many multiverse models about the history and origin of the universe, and I've seen people express concerns about the possibility of "infinite history" in much the same words you've used here. But that's a different issue than Smolin's notion of infinitely many layers of theory being needed to keep the cosmic machinery running.
Personally, I'm not overly bothered by the notion of space and time existing forever in both directions, with new bubbles of interesting reality occasionally spawning themselves from the remnants of the old ones at random after uncounted eons of darkness. I don't know why that structure might exist rather than something else (or why there is something rather than nothing), but positing a "first cause" wouldn't make me any more satisfied (and would raise additional questions). I think modern math and science are more comfortable with the possibility of infinity than folks used to be, and that our notion of a "cause" is rather different. (When I think of "causality", all I think about are light cones and Cauchy problems: http://en.wikipedia.org/wiki/Cauchy_problem .)
But eventually someone pointed out that before we understood the history and nature of the solar system, natural philosophers looked for similarly deep explanations for the orbits of the planets. Kepler once proposed a deep connection between the (five) gaps between the six known planets and the five platonic solids (see, e.g. http://www.pbs.org/wgbh/nova/blogs/physics/2011/12/beautiful...), and Bode's law for planetary orbits was widely accepted until Neptune was discovered in the "wrong" place in 1846 (see http://en.wikipedia.org/wiki/Titius%E2%80%93Bode_law).
Today, we understand that there is no reason to expect any particular pattern of planetary orbits: solar systems are a dime a dozen in the galaxy, and their details are accidents of history. So while I don't like the idea, I've gradually come to accept that the "fundamental physical constants" of our universe could conceivably turn out to be just as arbitrary as the ratio of Jupiter's major orbital axis to Saturn's.