Hi Scott, I'm sure Tom will address the rest in more details. Just one general comment:
I am currently assuming that the histogram matching does not use model building as part of it
This is a reasonable expectation since density modification and model building are two logically separate tasks that can benefit from each other in a number of ways but do not need to be executed all together. Density modification is a process of non-linear map transformation with the purpose to impose some prior knowledge on the map and thus improve its quality. Note: no model or model building is involved into the definition! PRISM approach may brake this logic, but you were specific enough with your question by mentioning "histogram matching", so this does not really apply then. Pavel