This study investigated the effects of the Mercedes Model with Embedded Assessment Strategy, subject specialisation and gender on students’ knowledge of selected biology concepts. A pretest post-test control group quasi-experimental research design was adopted for the study. The sample consisted of 666 Senior Secondary Class II biology students from six senior secondary schools randomly selected from two purposively selected Education Districts (EDs) in Lagos State. The two EDs were randomly assigned to treatment such that one was the experimental group and the other the control group. Seven null hypotheses were formulated to guide the study. Also, five instruments were used to generate data for the study, namely, the Mercedes Model with Embedded Assessment Strategy in Diffusion (MEASID), Mercedes Model with Embedded Assessment Strategy in Osmosis (MEASIO). Conventional Lesson Plan on Diffusion (COLPOD), Conventional Lesson Plan on Osmosis (COLPO) and Test on Students’ Learning Outcomes in Osmosis and Diffusion (TESLOOD). Data collected were analysed using mean, standard deviation, analysis of covariance (ANCOVA) and multiple classification analysis. The study has shown that the use of the Mercedes Model with Embedded Assessment Strategy has positively improved students’ knowledge of selected biology concepts. Further, the biology students in the science class were better than the non-science group at imbibing the knowledge of biology concepts. In the same vein, the males performed better than their female counterparts in knowledge. It was recommended, among other things, that teachers of biology adopt the Mercedes Model with Embedded Assessment Strategy in the teaching of students in secondary school.