That would involve a much, much, much more complicated simulation than the one Peter Norvig is writing about.

Yes it would, though that kind of modelling is some orders of magnitude outside of the scope of the toy model in the article.

This is way to simple a model for that - it's a very nice toy model, which is useful to get a quick intuition, but you cannot just add pieces to it and hope to get meaningful results.

A colleague of mine is working on simulating the banking system according to the real Basel regulatory framework rules.

http://www.iiim.is/2012/10/threadneedle/

Such a simulation would reveal a lot about the creator's beliefs, and essentially nothing about reality.

Making the creator's beliefs (including hidden assumptions) completely clear is valuable all by itself. But in addition to revealing the creator's beliefs it also reveals arithmetic errors, which are quite common in poorly specified verbal models.

For an example of the latter, observe this HN thread: https://news.ycombinator.com/item?id=7042469

I'm surprised you didn't include a link to this thread, where I still owe you some work...

https://news.ycombinator.com/item?id=6725096

Yeah, but that's a discussion I didn't want to get into again. A simple, completely indisputable arithmetic error made the point better.

You may find this interesting:

http://www2.econ.iastate.edu/tesfatsi/WhyEconomistsShunSimul...

That's all any model used to construct a predictive simulation can do.