Appearing recently methods, having guilds of artificial intelligence, permit on building of simulating models which realize assigned tasks on the basis of patterns taken directly with nature observation [1]. The processing techniques based on artificial neural networks create a special group, being in fact a computer simulators of brain work [3]. With the help of neuronal models it is possible to predict the expected crops yield on the basis of empirical data regarding crop yields in last summers. This work proposes utilization of prediction methods, which represent chosen topologies of neuronal nets among others, the RBF (Radial Basis Functions) neural network peculiarly.