Usual reductionism considers systemic, acquired properties as non-systemic, possessed properties. We consider here the non-systemic usages of systems, misunderstood as non-interacting virtual objects or devices, and the misunderstanding between non-complex (first Systemics) and complex systems (second Systemics) as another form of reductionism. This reductionism leads to inappropriate and ineffective approaches, particularly dealing with complex systems such as socioeconomic systems, whose complexity is often misunderstood and neglected. However, this reductionism should be distinguished from mixed usages of systemic approaches suitable to deal with multiple, dynamic, temporary, and partial systemic natures of phenomena related to complex systems. We consider that we should move from the well-defined, often simplistic, world of Systemics to Quasi-Systemics, which is intended as constructionist Systemics, always in progress, non-ideological, multiple, contradiction-tolerant, incomplete, and in its turn emergent. Rather than recommending a pragmatic attitude, we mention two approaches, one methodological approach called Logical Openness and another, the Meta-Structure approach, which is suitable to more formally deal with such multiple aspects and-based on mesoscopic representations-suitable to represent quasiness.