This special issue is devoted to describing and evaluating new methods for analyzing online writing processes in a second language (L2). As all the contributors have stated, the aim is to identify the cognitive processes underlying writing. The ultimate aim is to employ observations and experimental manipulations of these processes to build theories that can be used to help writers write more effectively in their second language and, as Lopez-Serrano, Roca de Larios, and Manchon (2019) argue, to facilitate learning of the second language. This project has two sides: (i) developing methods for recording and analyzing behavior as it unfolds during writing and (ii) relating this behavior to underlying cognitive processes. Furthermore, these two sides to the question are interdependent: what behavior is recorded and how it is analyzed depends on the researcher's theory of the cognitive processes involved, and, to a certain extent, theories of these cognitive processes depend on what it is possible to observe. In this commentary on the articles in the special issue, we will first consider the theoretical models within which the research has been framed. We will then consider the contribution of the articles to developing methods for observing the writing process, and how these have been related to underlying cognitive processes. Finally, we will consider directions for future research and some of the possible implications for second language writing instruction.