We used data from the Midlife Development in the United States (MIDUS I) (N = 2,031) to compare three models of how work-family conflict and enrichment might operate to predict well-being (mental health, life satisfaction, affect balance, partner relationship quality). We found no support for a relative-difference model in which the conflict-enrichment balance predicted outcomes. In the work-to-family direction, the additive model fit best: Both work-to-family conflict and work-to-family enrichment were independently linked to outcomes. In the family-to-work direction, the interactive model fit best: Family-to-work enrichment buffered the negative outcomes ordinarily linked to family-to-work conflict. Enrichment is key because with the additive model, it contributed incremental explanatory power, and with the buffering model, it conditioned conflict-outcome relationships. Work-to-family conflict and family-to-work enrichment appeared particularly salient for well-being.