This paper addresses quantitative analysis and performance improvement of feedbacked passive systems. Matrix-parameters are integrated into the passivity to enable performance evaluation of passive dynamical systems. With the matrices, the feedback system composed of the passive subsystems is quantitatively evaluated. It is further shown that the performance of the feedback system is improved as compared to that of the disconnected subsystems. Then, the quantitative analysis of the feedback system is extended to that of an iterative feedback system. Motivated from the analysis of a decentralized controlled system, we assume that each subsystem in the iterative feedback system has a special structure. Then, the gradual performance improvement is shown via the additional subsystem connection.