In the above paper,(1) an interesting problem of learning motion trajectories with different time scales based on the learned control inputs was considered. However, the author was apparently unaware that such an idea was previously reported in [1], [2] based on the time-scale property of robot dynamics [3]. In this note, we briefly review the literature and point out the difference between the works in and [1], [2]. We shall show that the proof developed in [1], [2] can also be applied to derive the time-scale learning controller far the class of nonlinear system considered in Xu.