Over a quarter century has passed since theoretical debates surrounding competing assimilation models emerged, and durably structured research on immigrants and their descendants in America and beyond. In this article, I offer a three-pronged reflection on the contemporary state of assimilation research. First, I aim to take stock of the relative merits of segmented and neoassimilation theories and their ability to explain major empirical trends in both the US and Europe. I argue that there is now an emerging empirical consensus about the second generation in the US and Western Europe primarily experiencing intergenerational progress rather than downward assimilation as envisioned by segmented assimilation theory. I then note six analytical challenges facing further theory building: clarifying the role of race and better understanding how cultural difference shapes assimilation trajectories, rethinking the relationship between immigrant socioeconomic mobility and the experience of belonging, acknowledging the importance of immigrant selectivity in conditioning assimilation, facing issues of in- and out-of-sample selectivity due to processes endogenous to assimilation such as ethnic attrition and incarceration,and studying the third generation. Looking out and into the future, I note the need for conversations across methodological traditions and specialist subfields to encourage further theoretical progress and assess existing data infrastructures and future data requirements. Finally, in tandem with machine learning applications allowing for empirical surprise and abductive reasoning, I argue that the current era of data plenitude and unprecedented ability to collect high-dimensional surveys through nonprobability online samples is likely to lead to further theoretical progress.