College students experience great stress due to many factors, such as an uncertain future, academic responsibilities, and pressures imposed by social communication. Many institutions of higher education are focusing on how to mediate stressful situations and increase the subjective well-being (SWB) of students to sustain a lifestyle focused on wellness. The online survey used for this study focused on testing an exploratory SWB model by implementing partial least squares structural equation modeling (PLS-SEM) techniques. The participants were 470 college-aged students enrolled in seven different institutions in six cities across China. Findings yielded a good model fit (the Standardized Root Mean Squared Residual [SRMR] = .054) with the validity of manifest variables, reliability of the latent variables (LVs), and overall SWB model indicating moderate predictiveness (GoF R2 = .476) by the LVs. Additionally, all of the direct path coefficients and indirect path coefficients that consisted of four partial mediators and one full mediator yielded statistically significant results via bootstrapping. Furthermore, path coefficients for utilization of emotion to life satisfaction for the cognitive exercise group were significantly higher than for the non-cognitive exercise group. The findings illustrated a good model fit for an exploratory SWB model that can predict an individual’s SWB, and cognitive and non-cognitive exercises had different effects on the individuals’ SWB.