This technical note studies the distributed coordinated tracking control for multiagent systems with model uncertainties. Both unknown model parameters and unknown system dynamics are considered. It is assumed that there exist parametric uncertainties and unknown dynamics with the informed agent as well, and only the state value of the informed agent can be accessed by a limited number of agents. With the utilization of neural network approximation and adaptive estimation, a new distributed adaptive tracking control is proposed to make all agents cooperatively follow the desired trajectory specified by the informed agent. The control design is first presented for the first-order multiagent systems, and then extension is made to the second-order multiagent systems using backstepping. A unique feature of the proposed control is that the unknown bounds of neural network approximation errors are also estimated online. Using Lyapunov stability theorem, it is rigorously proved that asymptotically cooperative tracking can be achieved under the assumption that the sensing/communication topology among agents is connected. Simulation results are included to illustrate the proposed control.