darkblu, you miss the point on two counts.
First of all, no one is denying that QM, string theory, Quantum Gravity, Theory of Everything, or whatever is ultimately at the lowest level (if it even exists) is what drives everything. It's all a mathematical model anyway, and the map is not the territory. But humans have developed mathematical models at several levels of reality to deal with aggregate and statistical interactions between very large numbers of particles, or elements. We don't do economics by studying particle physics and attempting to predict people's economic behavior. And we don't do macroeconomics by predicting everything from microeconomics.
When humans design things, we use the level of mathematical model appropriate for the task. When we design buildings, we use newtonian mechanics. QM is still there, but any QM specific effects are lost in the noise. When we design circuits, we are now starting to have to use/be aware of QM (because of the size reductions). When we design airplanes, we base it on fluid dynamics. The question is, at what level does the brain work?
To follow up with my second relevant point: The theory of computation was designed using abstract mathematics, not physics. We constructed several differenct types of "devices", such as Finite State Automata, and Turing machines, all without any reference to the laws of physics. These are mathematical/logic constructs like PI. Several conceptial models machines are in fact impossible in reality.
Up until recently, all of our physical devices that do computation are essentially variations on these abstract models. In Automata theory, we consider the Turing machine to be "universal" meaning that it can compute anything that is computable (e.g. brain is just a machine, in theory, a Turing machine could compute all the inputs and outputs of the neurons and simulate a brain). The Church-Turing thesis conjectures that there is no machine that is more powerful than a Turing machine. And for the last few decades, try as people might, they could not come up with any new construct that can compute problems that a Turing machine couldn't. In fact, they could not even construct a machine that could compute problems asymtotically more efficient than a Turing machine. (I'm leaving out non-physically realizable non-deterministic models)
That was the case, until Peter Shor came up with a way to factor numbers using qubits. The best known algorithms for factoring numbers are exponential. Meaning, on any classical device, the time needed to factor grows exponentially with the size of the number. It is not known if there is a more efficient classical algorithm, but people suspect not (and modern cryptography depends on it)
So Quantum factoring contradicts the widely held notion that classical computation was the best you could ever do by solving the problem in polynomial time.
This means that computers designed to take advantage of macroscopic reality (classical design) can solve a class a problems wholly different than computers designed to utilized microscopic quantum effects.
Now there are two possibilities: Either the Church-Turing thesis is right and there is a fast classical algorithm for factoring, OR, Quantum computers represent a new class of computation model and Church-Turing was wrong.
The way this fits into the brain is this: Does the brain operate according to "classical" models, or does it utilize qubits and other weird quantum techniques, such that it is impossible to simulate a brain on a classical device, but could be simulated on a quantum device. I assert that biology appears to be classical, e.g. DNA, ribosomes, all stochastic classical processes.
This is the difference you are failing to recognize. The fact that classic devices (like your Intel CPU) operate at the lowest levels of reality on Quantum Mechanics, doesn't mean that they are Quantum Computers. They are in fact, variations on finite state automatons and ignore quantum effects that they could benefit from.
The fundamental assumption in classical computation models is that at each time step T, only a finite amount of "work" can be done. Quantum computing shows that a near infinite amount of work can be done at each step (by utilizing the power of the machine in multiple universes, I prefer).
A device designed to ignore quantum effects will never have that power, even though the device itself exists in a universe that is quantum.
