The testing effect is both robust and generalizable. However, most of the underlying studies compare testing to a rather ineffective control condition: massed repeated reading. This article therefore compares testing with note-taking, which has been shown to be more effective than repeated reading. Experiment 1 is based on a 3 × 3 between-participants design with the factors learning condition (repeated reading vs. repeated testing vs. repeated note-taking) and final test delay (5 min vs. 1 week vs. 2 weeks). It shows that in the immediate condition, learning performance is best after note-taking. After 1 week, both the note-taking and the testing groups outperform the rereading group, and after 2 weeks, testing is superior to both note-taking and rereading. Since repeated notetaking may not be the most effective (and common) operationalization of note-taking, Experiment 2 contrasts repeated testing with 2 other note-taking conditions: note-taking plus note-reading and note-taking plus testing (with only a 2-week final test delay). Both conditions that include a testing phase result in better long-term learning than note-taking plus note-reading. In summary, our findings indicate that—in the long run—testing is a powerful learning tool both in isolation and in combination with note-taking.