Thomas Kuhn’s theory of paradigm reveals a pattern of scientific progress, in which normal science alternates with scientific revolution, but he underrated too much the function of scientific test in his pattern. Wesley C. Salmon pointed out that, on criticizing the so-called testing pattern of science, Kuhn focused all his attention on a single testing model, namely hypothetico–deductive (H–D) schema. However, as a matter of fact, many philosophers of science had already abandoned that schema and taken Bayesian schema as a proper testing model. The main difference between Bayesian schema and the H–D schema lies in that the former is a testing model for more than one theory while the latter just for a single theory. Since Kuhn, multi-theoretical testing model has become aconsensus among experts, that is, a theory and its rivals should be faced with testing together, rather than a theory being tested in isolation. Kuhn was correct in finding the H–D schema not appropriate to scientific test, but didn’t catch the propriety of Bayesian schema in this field. This led to his disapproval of the logic or method of scientific test. I agrees largely with Salmon’s appraisal of Kuhn’s view on scientific test, and gives a further argument for it. I’ll employs Bayesian schema to re-examine Kuhn’s theory of paradigm, uncover its logical, or rational, components, and thereby illustrate the tension structure of logic and belief, rationality and irrationality, and comparability and incommensurability in the process of scientific revolution.