In this paper, a general adaptive control design aiming at asymptotic tracking is developed for NARMA (nonlinear auto-regressive moving average) systems with periodic time-varying parameters and periodic unknown time delays in the nonlinear uncertainties. Various challenging issues like periodic varying parameters, periodic time delay and even nonlinear nonparametric uncertainty are addressed in this work with nontrivial efforts of integrating several novel ideas emerged in recent years. The asymptotic tracking performance and closed-loop stability are rigorously established by theoretic proof and verified by simulation studies.