Motivated by declining fish populations and the apparent inability of regulatory agencies to manage important fisheries, this research measures the accuracy of a fishery model that explicitly models the regulatory process and the resulting degree of compliance by fishers.  The method involved careful review and enhancement of a prior model with a more limited regulatory sub model,and then measuring, for both models, the mean absolute error of model calculated values for historical spawning biomass, acceptable biological catch, and harvest.  The most recent five years of data were held back so that model prediction error could also be computed.  Results indicated that although the fitness error for the enhanced model was significantly less than the prior model (23% vs. 38%), predictions were improved only for one of the three measures.  The implications for researchers seeking to endogenously model fishery management processes are sobering. Policy makers on the other hand will likely see the results as support for their instinctual skepticism regarding policy models.