Constructive neural network (CoNN) algorithms enable the architecture of a neural network to be constructed along with the learning process. This chapter describes a new feedforward CoNN algorithm suitable for multiclass domains named MBabCoNN, which can be considered an extension of its counterpart BabCoNN, suitable for two-class classification tasks. Besides describing the main concepts involved in the MBabCoNN proposal, the chapter also presents a comparative analysis of its performance versus the multiclass versions of five well-known constructive algorithms, in eight knowledge domains, as empirical evidence of the MBabCoNN suitability and efficiency for multiclass classification tasks.