moltenmetal and 2dye4 - I am going to make a wild-a$$-guess here, but it sounds to me like you don't work with numerical simulations (FEA, CFD, etc) in your working career. These are the types of numerical simulations that we are talking about with the "climate models". cvg hits the nail on the head in that regard. These models discretize the simulation space (on our case, the atmosphere, the surface, and by some measure the oceans). In each discretized volume (if I were creating a climate model, I would use a control volume approach, rather than a Galerkin-method discretization), there is stuff that goes on inside the cell, and there is flows across the boundaries: both energy and mass transport terms. A reasonable model should be able to simulate such large-scale features as Hadley Cells, the Jet Streams, the Antarctic Polar Vortex, etc. However, small-scale features such as thunderstorms (and even smaller, such as clouds) are not directly simulated, but are somehow accounted for by what I can only describe as fudge-factors. These models are started with a set of initial conditions and assumed boundary conditions (volcanoes, CO2 concentration change assumptions, etc). Then, these simulations march off in time.
Sure, the models have F=ma, temperature/specific heat capacity relationships, buoyancy, conservation of mass, conservation of energy, etc. Those are the fundamental equations also included, but are not the sum of the "model". The algorithms have assumptions (clouds, for example) that, while perhaps based on a relationship between temperature at height, RH, etc, is still not well understood. We don't even have a fundamental understanding why clouds form discrete entities...
If you don't work with numerical simulations, I can understand why you may put some measure of faith in them. To almost a person, everyone that I know that works with numerical simulations is rather critical of the "climate models". Sort of a situation of "it takes one to know one".
And don't even get me started on discretization error...