This study aims to determine the effectiveness of the project-based learning (PjBL) model compared to the cooperative learning (CL) model in improving the critical thinking skills and chemistry self-efficacy of class 10 students simultaneously and consecutively. This quasi-experimental study with a post-test-only control group design was conducted at Public Senior High School in Yogyakarta. The research sample consisted of 65 students from an experimental group of class 10 A 4, totaling 33 students, and a control group of class 10 A 1, totaling 32 students, obtained through cluster random sampling techniques. The instruments used were critical thinking ability tests and self-confidence questionnaires. Before being used, the instruments were tested for content and empirical validity, and normality tests and multivariate outlier tests were carried out, followed by unidimensional assumption tests, parameter invariance, and local independence. The experimental group was taught using the PjBL model, while the control group was taught using the CL model. Critical Thinking Skill Fest (CTST) and Chemistry Self-Efficacy Scale (CSES) were applied as post-tests only to students in both groups. One-way multivariate Analysis of Variance (MANOVA) and post hoc test between subjects (Tukey Test) showed that applying the PjBL model was significantly better than applying the CL model on students' critical thinking skills and chemistry self-efficacy simultaneously and individually. The effectiveness of the PjBL model on students' critical thinking skills and chemistry self-efficacy simultaneously and individually is in the moderate category compared to the CL model. This research provides actionable insights into the advantages of implementing PjBL to improve critical thinking and self-efficacy among students. This research can serve as a reference for future research exploring the effectiveness of PjBl learning models in other subjects or educational levels.