This paper deals with global synchronization in arrays of coupled delayed neural networks with delayed coupling. Through employing Lyapunov-Krasovskii functional and Kronecker product technique, one novel synchronization criterion is presented in terms of linear matrix inequalities(LMIs) based on the integral inequality and convex combination, in which the condition is dependent not only on lower and upper bounds but also on delay derivative. Through employing Matlab LMI Toolbox and adjusting some matrix parameters in the derived results, we can realize the design and applications of the addressed coupled systems. The applicability of the proposed methods can be demonstrated by one numerical example with simulations.