The physics learning process at school does not involve students directly which causes the low cognitive learning outcomes of students. For this reason, it is necessary to apply an effective learning model for improving student learning outcomes. One suitable learning model is the treffinger learning model. The treffinger model provides learning activities with three stages namely basic tools, practice with the process, and working with real problems. These three stages can make students actively and creatively involved in the learning process. The purpose of this research is to find out the effectiveness of increasing cognitive learning outcomes of students on optical material by applying the Treffinger learning model. This type of research is a quasi-experimental design with a non-equivalent control group design. The research sample was taken two classes randomly as the experimental class and the control class from the student population in class XI IPA of SMA Negeri 9 Pekanbaru. The data collection instrument used students' cognitive learning achievement test consisting of pretest and posttest given to the experimental class and the control class. Data were analyzed using descriptive and inferential analysis. Descriptive analysis results obtained an average absorption capacity of 72.64% experimental class with a good learning category and 63.47% control class with quite a good category. With the results of the inferential analysis with an independent-sample t-test, there are significant differences in learning outcomes between the experimental class and the control class, where the experimental class has a higher increase in learning outcomes. Thus the use of the treffinger model is effective in improving student learning outcomes in class XI IPA of SMA Negeri 9 Pekanbaru on optical material.