The　problem　of　command-filter-based　adaptive　tracking　control　is　investigated　for　a　class　of　stochastic　nonlinear　systems　with　strict-feedback　structure　with　input　dead-zone　in　this　paper.　Radial　basis　function　neural　network　(RBF　NN)　is　employed　to　approximate　the　packaged　unknown　nonlinearities.　In　order　to　eliminate　the　influence　of　＇the　explosion　of　complexity＇　which　will　exist　in　the　conventional　controller　design　process　via　backstepping　technique,　the　control　method　of　the　command-filter　is　introduced.　For　the　problem　of　input　dead-zone　which　appears　in　the　stochastic　nonlinear　systems,　which　will　be　dealt　by　a　reasonable　method,　namely,　the　dead-zone　nonlinearity　can　be　regarded　as　a　combination　for　a　linear　term　and　bounded　disturbance-like　term.　Combined　adaptive　backstepping　design　algorithm　and　Lyapunov　stability　theorem,　an　adaptive　neural　command-filter　controller　is　developed.　The　proposed　control　scheme　reduces　the　calculation　burden　due　to　the　repeated　differentiation　for　the　virtual　control　laws　and　guarantees　all　the　closed-loop　signals　remain　semi-globally　uniformly　ultimately　bounded　(SGUUB)　in　the　sense　of　the　four　moment.　And　the　tracking　error　converges　to　a　small　area　near　zero.　Meanwhile,　the　effectiveness　of　the　presented　approach　is　proved　by　simulation　results.