Data processing applications are increasingly heterogeneous, mixing different technical domains such as telecommunications, digital and analog electronics, signal processing algorithms, etc. These domains have different methods of modeling and design that consider their components and relationships in different ways. To mix these different domains, modeling languages and platforms generally use a hierarchical approach where the hierarchy of the model is coupled to the change of model of computation. Moreover, heterogeneous components cannot be used and what happens when data crosses the boundary between two domains depends on the modeling environment. This paper presents a flat heterogeneous modeling approach that solves these issues by using Heterogeneous Interface Components, which gives more control to the designer on the semantics of the interactions between models of computation.