We suggest new directions for research on the effectiveness of concurrent product development. These models relate overlapping or functional integration to performance using analytical, simulation, or statistical methods. By collecting quantitative data from 82 research articles, we ascertain the percentage of a methodology's models that have incorporated each of certain significant theoretical features. We then use this information to recommend important underutilized features and a comprehensive theory that integrates the features. Analytical and simulation modelers could make more use of functional integration, while statistical modelers could attend more to overlapping, rework, and the constituent phases of the development process. Then, in finding a subset of interrelated features appearing simultaneously in a majority of a methodology's models, we identify a core theory upon which models in the mainstream build. Next, our study offers specific mutual learning suggestions through which modeling efforts in any one methodology might facilitate modeling efforts in another methodology. Analytical and simulation modelers could study the reluctance to transmit and use preliminary information, while statistical researchers could develop dynamic models. Finally, the study provides a template for researchers willing to apply our evaluation methods to other areas.