This study investigates, through a cluster-randomized trial, the effectiveness of two approaches to increasing middle school students' science learning using a traditional science curriculum. Ninety schools were randomly assigned into one of three arms: (a) a treatment arm in which the textbook curriculum was modified based on four principles of cognitive science coupled with teacher professional development (PD), (b) a second treatment arm in which teachers received PD designed to improve their knowledge of the science content, and (c) a business-as-usual control group. The PD was able to change teacher practice but barely improved teacher knowledge. No significant boost in student achievement was observed, except in a few instances, where there were some promising findings. Exploratory analyses were conducted to examine what makes the interventions more effective. Implications for future research were discussed.