Wonder if you read the OP's request before commenting? He was referring to getting into the quant stuff that is discussed on the Wilmott forums. I gave him an apropriate answer. I make no assertion about the merits of what he wants to achieve or the relevancy to his trading. That's for him to decide.
Perhaps. Once the OP has learnt enough math to be able to decipher the notation and language used on the Wilmott forums, he/she may realize it's not what they were looking for. Then again, it may be exactly what they were looking for. Perhaps the OP wants to automate some stat arb? The stochastic calculus probably isn't going to help there but the linear algebra just might. Who knows?
Forget about math models. Remember LTCM?, complex models by four nobels meant blowup. (and they even owed big money)
From Joshi's list of recommended reading: LOL. Btw, I have one of his books, it's very dry (forget the title at the moment).
Yes. First, figure out the problem(s) you're tackling, then go find the tools and skills you need for the solution. The OP is looking in the wrong end of the telescope by reading a pile of math books without any idea of their relevance to his subject domain. Not a fuel-efficient way to run your brain.