Based on the principle of sliding mode control and property of Nussbaum-type functions, an improved adaptive neural control scheme is proposed for a class of MIMO nonlinear time-varying delay systems with unknown function control gains. By choosing appropriate Lyapunov-Krasovskii functionals, unknown time-varying delay uncertainties are compensated for. In this paper, the restriction of control gains is relaxed, and the system considered here is more general. By utilizing Young's inequality, the assumption of time-varying delay uncertainties is relaxed. Moreover, only one parameter is adjusted in each subsystem, and the complexity of implementation is reduced. By theoretical analysis, the closed-loop control system is proved to be semi-globally uniformly ultimately bounded. Finally, simulation results are provided to illustrate the effectiveness of the proposed approach.