I received a friendly mail from John Hallam about my Archean project:
Off the top of my head, what you have described sounds like an Euler-method integrator for the laplace equation with 6 coupled potentials. You should be able to get lots of interesting behaviour out of that. If you make the local kernel even more flexible -- rather than just an average, which I think implements the laplace operator -- you could allow other local operators instead -- then you have a system that approximates the solution of other PDEs, for example the wave equation,
Unfortunately I'm not really sure what this all means. I can see how Archean is like systems that predict the spread of diseases, which I think the Eurler-method is used for. I would be interested if someone could point me towards more local operators.
The people who really run organizations are usually found several levels down, where it is still possible to get things done.