The NK model is designed to study evolutionary adaptation in rugged fitness landscapes. The factor $K$ determines the number of interacting genes, their degree of pleiotropy and epistasis, and consequently the ruggedness of the fitness landscape. However, in natural organisms, the degree of epistatic interactions and the number of functions a gene can have are to a certain degree determining the ruggedness of the landscape. Still, pleiotropy and epistasis can evolve independently from each other, and are to some degree independent of the ruggedness of the landscape. Here, we propose an extension to the standard NK model to investigate these factors independently of each other. Over the course of evolution the computational model organisms can now change how their genes interact and how they control phenotypic traits. Further, the degree of epistasis and pleiotropy is affected by the ruggedness of the landscape and becomes reduced with increasing ruggedness. While this proves that the extension of the model performs as expected, the adaptations are minor, presumably because only relatively short periods of adaptations with few mutations can be studied.