In the mentioned paper(Tan, 1996, Automatica, 32, 1701-1706) a nonlinear one-step-ahead control strategy based on a neural network model was proposed for nonlinear SISO systems. In this correspondence it is pointed out that the control law given by Tan lacks of reasonable explanation, thus the conditions of output stability and convergence are incorrect as well. An alternative control algorithm and the its stability analysis are suggested.