Due to the ever increasing demand for high-performance computers, parallel architectures have emerged as a strong research area. Most of the investigations in this area are based on the conventional Von-Neumann (control flow) model and the more radical data flow approach. The data flow approach, while being conceptually attractive and a possible solution to the problems encountered in parallel Von-Neumann architectures, is nonetheless plagued with problems of its own. Architectures which attempt to combine two or more models are beginning to appear as alternatives. A discussion is presented of preliminary ideas on combining data flow and control flow in a single architecture.