In this study, finite-time synchronisation between two chaotic systems with discrete and distributed delays is investigated by using periodically intermittent memory feedback control. Based on finite-time stability theory, some novel and effective synchronisation criteria of intermittent control are derived by means of linear matrix inequalities and differential inequality techniques. Furthermore, a necessary condition of finite-time synchronisation of intermittent control is given for neural networks with discrete and distributed delays. A numerical example on two chaotic neural networks shows the effectiveness and correctness of the derived theoretical results. In addition, a secure communication synchronisation problem is presented to demonstrate practical effectiveness of the proposed method.