Because fuzzy neural networks technique has the advantages not only strong adaptability to nonlinear systems but also quick learning ability, it has been applied to nonlinear systems successfully. an optimal terminal missile guidance law based on fuzzy neural networks for the advanced missiles with thruster vector control is presented. In the design of optimal guidance law, the minimum of missile energy loss and miss distance is required while the nonlinear character of the thruster vector control is considered. In addition, fuzzy learning rules are presented to improve the learning ability of the neural networks system. Numerical simulation results are given to illustrate that the presented fuzzy neural networks guidance law has good performance to hit maneuvering targets.