Indeed. To respond to another early posting, the uniformity was actually a bit unexpected. The best explanation, for reasons you can read elsewhere (e.g. Wikipedia's excellent articles on the universe), is that the universe expanded very, very rapidly during the first very, very small fraction of a second. While it's true that the universe is currently expanding, that expansion is nothing compared with this initial period of "inflation." The passage quoted about God stretching out the universe would in fact fit quite nicely with this initial period of inflation, since that inflation could reasonably be described as a stretching of space-time.
Well, this is perhaps being a little bit pedantic of me, but the level of variation
is right around what was expected, not because of arguments regarding what was going on in the early universe, but instead because it had to be around the level of what was observed in order to explain the galaxies and galaxy clusters that we saw. Granted, our error bars were quite large before we launched the COBE satellite, but they couldn't have been different by more than about an order of magnitude.
The issue that you're talking about is a little bit different. If you simply take the standard big bang theory, with normal matter, radiation, and a cosmological constant, then that theory cannot explain
any level of uniformity in the early universe. The uniformity is an input to the theory, but it actually predicts that different parts of the sky
have never been in contact with one another. Inflation is a proposed solution to this because it modifies the early expansion history, bringing everything that we observe into contact in the early universe.
Most people think inflation very likely today because inflation not only explains why different parts of the universe actually were in contact in the past, but also provides a physical mechanism for producing the seeds for the small deviations from perfect uniformity. And those deviations from uniformity, the parts on the CMB that are warmer or cooler at the 0.001% level show the exact sort of pattern we expect from inflation.
There are, however, other proposed explanations. For example, loop quantum cosmology attempts to explain the early expansion history not by proposing a new component of the universe, but instead by proposing a new theory of gravity (loop quantum gravity) and estimating how that theory modifies the early expansion. The ekpyrotic (meaning "conversion into fire") theory is another alternative which sees our universe as being stuck on a sort of sheet in a higher-dimensional space (this is based upon string theory), and the hot early universe was produced when the sheet (or brane) containing our universe collided with another, heating up both branes to a very high temperature across the entire surface of the collision basically at the same time.
The hope is that in the next few decades we can distinguish between these theories through measurements of the polarization of the CMB. Planck is going to try to measure the polarization, but we'd have to be really
really lucky on the level of polarization in the CMB for Planck to see it.