This study explores the factors affecting cyberbullying behaviour through their attitudes and intentions based on a model developed using Social Cognitive Theory and the Theory of Planned Behaviour. Results from a survey of 428 Malaysian tertiary students were analysed using a two-step Structural Equation Modeling (SEM)--Artificial Neural Networks (ANN) approach. The SEM findings indicate that the intention to engage in cyberbullying is the most influential factor in cyberbullying behaviour, shaped by factors such as image (0.492), moral disengagement (0.211), perceived behavioural control (-0.141), university climate (-0.097), subjective norms (0.662), peer to peer relationships (0.108) and attitude (0.235) towards cyberbullying. The ANN results show that image is the biggest predictor (36.35%) of cyberbullying intention, followed by moral disengagement, cyberbullying attitude, PBC, and university climate. These insights provide a deeper understanding of the underlying factors behind cyberbullying among Malaysian tertiary students and offer guidance for reducing cyberbullying in the country.