The development of maximum power point tracking (MPPT) is continuing in order to increase the energy transfer efficiency of the solar photovoltaic system. This paper provides a review of the conventional maximum power point tracking techniques that is enhanced by the presentation of a new technique. The new method is based on a genetic neural algorithm in order to predict the closest point to the maximum power point (MPP), which will be the kickoff point of the search process. Not only does the new technique start the search process from the nearest point to the MPP, but also the developed search algorithm is very fast. Consequently, the time taken to reach the MPP is reduced. In order to determine the new MPPT performance, a complete photovoltaic generator system is modeled and simulated using the MATLAB/SIMULINK package. Simulation results show that the new technique reaches the MPP in less than 100 sample times compared to tens of thousands of samples for conventional methods. Furthermore, the new technique reaches directly the target MPP with small deviation from the intended values. Consequently, the new technique has a significant improvement in energy extraction efficiency from the photovoltaic array to the load, in addition to higher tracking speed and system stability compared to the conventional ones.