Dynamically evolutive models and recursive equations for self, antigen, dynamic computer forensics, immune tolerance, mature-lymphocyte lifecycle and immune memory are presented. Following that, a new model, referred to as Insdcf, for computer network surveillance and dynamic computer forensics is proposed. Simulation results show that the proposed model has the features of real-time processing, self-learning, self-adaptivity, and diversity, thus providing a good solution for computer network surveillance and dynamic computer forensics.