Motivated by capturing more test-takers’ problem-solving behaviors and further providing richer diagnostic feedback, technology-enhanced diagnostic assessments (TEDAs) are becoming increasingly popular. This study proposed an innovative cognitive diagnosis model for analysis multisource data collected from TEDAs, including outcome data, process data, and biometrical data. The proposed model can be seen as an extension of joint cognitive diagnosis models for item responses and response times by incorporating additional biometrical data, i.e., visual fixation counts (VFCs), an essential eye-tracking indicator to reflect the degree of attention a test-taker invested in the test. The feasibility of the proposed model was examined using simulated data. Results mainly indicated that utilizing the extra information from VFCs not only improves the model parameter estimation accuracy but also additionally provides the diagnostic results about the degree of attention a test-taker invested in the test. An illustration of the model by means of an application to real data was presented.