The emerging field of plasma medicine opens new therapeutic opportunities with the use of Cold Atmospheric Plasma (CAP) as a versatile tool for the treatment of tissues in various medical indications. Yet, the complexity of this very reactive medium combined with a high dependence on its environment and generation parameters make it difficult to predict and optimize such treatment. To this end, a simple yet robust and accurate tissue phantom allowing to study the penetration and distribution of plasma action in simulated in vivo conditions has been developed by several groups. It combines a hydrogel-based matrix closely resembling tissues and chemical reporters incorporated in the gel to measure the delivery of Reactive Oxygen and Nitrogen Species (RONS) by the plasma. This paper reports the use of these models in the literature to give an overview of the state of the art, their capabilities and the further research required to improve it. First the hydrogels composition (i.e. gelatin and agarose) is discussed, as well as the parameters allowing to fine tune the model. In particular, we show that modifying mass fraction has been reported to mimic several types of tissues and that different model configurations allow to test different treatments conditions, including the barrier effect of skin or the direct treatment of a tissue bulk. The role of other critical parameters is highlighted, including manufacturing, diffusion, electrical characteristics but also liquid composition, thickness, aging and temperature influence. Secondly, RONS reporters used in the plasma medicine literature (colorimetric/fluorometric dyes) are summarized. The analysis techniques are discussed and the dyes characteristics (i.e. wavelength, specificity, concentrations) are reported. Finally, the influence of medium and time on these measurements are covered. For both sections, the limitations of these current models are presented and linked to potential improvements and further research.