Important sources of disturbance torque inputs, which may severely degrade the seeker tracking accuracy of a guided missile, involve seeker mass imbalance, seeker gimbal friction, and seeker head restoring torque. The paper presents an approach to treat the disturbance rejection problem for the missile seeker. As the disturbance variables are measurable, we show that the quality of seeker tracking accuracy can be improved by the addition of feedforward control. A multilayer feedforward neural network is proposed to realize the nonlinear adaptive feedforward controller. Extensive simulation studies are performed to examine the effect of the proposed control scheme. The result shows that the neural controller affords possibilities for improving the miss distance performance.