This paper addresses the output tracking control problem and lumped disturbance rejections for nonlinear cyber-physical systems described by fuzzy model systems with parametric uncertainties, external disturbances and cyber attacks. Especially, to assure both the robust tracking and lumped disturbance estimation performances, an uncertainty and disturbance estimator (UDE)-based modified repetitive control (MRC) protocol is devised for the addressed control system. In detail, the impact of unpredictable factors, such as disturbances and uncertainties in the system model is robustly tackled by the UDE method with an appropriate filter. To be specific, the MRC protocol is amalgamated into the UDE technique for estimating the changes that occur in the afore-said unpredictable factors to ensure the asymptotic tracking performance of the system under study. Notably, to improve the tracking precision, the parallel distributed compensation strategy and the estimated lumped disturbance signal are incorporated into the control design. Meanwhile, the cyber attacks are depicted in this study as a stochastic variable that adheres to the Bernoulli distribution. Further, the required controller gain matrices are determined to affirm the robust asymptotic tracking performances with the aid of Lyapunov stability theory. Eventually, the method proposed in this paper is validated through the simulation results with two numerical examples, which include the continuous stirred tank reactor (CSTR) plant model. Further, the result reveals that the proposed control law provides the minimum tracking error when compared with an existing work.