I believe the PISA comparison in this context is flawed. To compare any statistics, we must control for other variables, otherwise any conclusion about causal effects and relationships could be deeply wrong. In the education context, the most important variables are the economic background of the households, the active involvement of the parents in the education of the children, the amount of time spending on the subjects (both in class and at home), the curriculum and methods of teaching.

Finland has only 5 millions population and a relatively high median income, the teacher:students ratio is much higher and the percentage of intact families are higher. Therefore, if we want to compare, we have to select better districts in the US, where the teacher:students ratio is similar high then see what the PISA scores will say. Only then, and looking deeper into the curriculum and teaching method, we can say whether homework is beneficent or not.