This paper reports research results on the designing of an intelligent controller by means of modeling human operator actions based on a Man-Machine Interaction Computer Experiment. Different disturbance signals are applied to the single and the double-integration systems under human operator control to collect data for training the Adaptive Network Based Fuzzy Inference-System as the Neuro-Fuzzy controller. The Neuro-Fuzzy system was able to replace the human operator even in the case of a random disturbance signal applied to the double integrator system.