Let's just say for now (and this is a significant simplification) that the Frame Problem is the question of how an organism separates semantically relevant information from irrelevant in a dataset which is so large as to prohibit exhaustive search strategies.
The microprocessor is basically a zero dimensional processor. It can only really manipulate one bit at a time. This means it is heavily serial. The brain, on the other hand is dimensional (parallel processing) but for the sake of argument, lets just say it is one dimensional. I'm not entirely sure about this, but my sense is that you could flatten out a neural network such that it was a richly connected one-dimensional mesh. Of course, it is debatable whether the brain can be treated as such a network, but this is irrelevant to the current discussion.
In addition to this spatial dimension, I think an additional dimension comes into play: time. The brain spreads processing across time. In language acquisition, for example, we don't build the simplest possible, most efficient storage structures like we do when writing software. We create rich structures which are organized according to their use. In the learning process, we are already doing pre-processing of future data, based on extrapolations from past experience. (see Cognitive Development, Language Development) Thus the brain is doing processing in at least two dimensions.
But this is interesting particularly because this dimension is theoretically available to a microprocessor "for free". A microchip could retain knowledge about past inputs and structure data according to how it will be needed/processed. This makes for a massively parralel computer--a one-dimensional processor. Since pre-processing (through organization) can be done every cycle of the computer's life, which is billions of cycles per second, the computer can effectively harness trillions of parallel processors as it matures.
This may require some sort of different architecture, and is perhaps related to my thoughts about integrating data and logic (programming) at the structural level (See Brains).