When designing artificial neural network (ANN) it is important to optimise the network architecture and the learning coefficients of the training algorithm, as well as the time the network training phase takes, since this is the more time-consuming phase. In this paper an approach to cooperative co-evolutionary optimisation of multilayer perceptrons (MLP) is presented. The cooperative co-evolution is performed on the MLP and training set at the same time. Results show that this co-evolutionary model reaches an optimal MLP with generalization error comparable to those presented by other authors but using a smaller training set, co-evolved with the system.