A methodology for process modeling, combining prior knowledge with neural networks (NN), and its uses within process control strategies are explored in this work The model type was the so called "Hybrid Neural Model" (HNM), based on fundamental conservation laws associated with a neural network used to model the uncertain parameters. Since a neural net within HNM has fewer parameters than a pure black bar NN model, an on-line training method may be used. The adaptive HNM approach was applied to two simulated processes : a highly non-linear CSTR and a four-stage flotation unit, The task of synthesis of HNM, its incorporation into MBC control strategies, and two an-line learning strategies are presented. Results obtained showed excellent performance and this approach can be considered an option in terms of flexibility and robustness for the control of complex processes.