Exponential　synchronization　for　a　class　of　delayed　discontinuous　complex-valued　neural　networks　(CVNNs)　with　leakage　delay　and　diffusion　effects　is　investigated　in　this　paper.　First,　CVNNs　are　separated　into　real　and　imaginary　parts,　an　equivalent　real-valued　subsystems　are　obtained　for　the　analysis.　Then,　some　novel　and　flexible　algebraic　criteria　are　established　to　ensure　the　exponential　synchronization　of　the　discontinuous　CVNNs　by　designing　feedback　and　adaptive　control,　which　improves　the　control　performance,　and　the　control　parameters　are　adjusted　automatically.　Compared　with　the　previous　research,　the　proposed　neural　network　models　are　more　practical　and　realistic,　and　the　obtained　results　take　into　account　the　impacts　brought　by　reaction　diffusion,　discontinuity　activations,　leakage　delay　and　complex　values　simultaneously,　and　the　research　enrich　the　previously　analysis.　Finally,　a　numerical　simulation　is　presented　to　verify　the　effectiveness　and　applicability　of　the　developed　theoretical　results.