The point is: The current proof of work scheme makes it possible to parallelize and have pools. A pool could thus become a very strong adversary which is not what we want - right? A non-parallelizable proof of work scheme has the consequence that nobody can become stronger than a, say, 4.5 GHz overclocked single core pentium. This is what we want.
It is possible to design a proof-of-work scheme where being part of a pool doesn't provide any (or negligible) advantage, aside from the benefit of having a more predictable revenue stream. Perhaps that's what the OP is wondering about.