Since 1940 the use of synthetic pesticides has led to considerable progress in agriculture and human health. In particular synthetic pesticides were used to protect crops and to fight against disease vectors. As a result it has been possible to feed most of the world population by increasing yields. Beside the beneficial effects for farmers by making their work easier and reducing harvest losses; and benefial effects for humanity by providing abundant food with improved sanitary quality, the intensive use of pesticides has given rise to serious health issues. Indeed pesticides can be very toxic and are responsible of farming diseases such as cancers and neurodegenerative diseases. Besides, with the increase of their efficiency and their selectivity, pesticides become also more and more expensive for farmers. However, in developed countries, there is a rapid change from subsistence farming to intensive farming, which is able to feed more people. In the past the regulatory framework for pesticide use was less restricting and this led to cases of abuse. In addition, our societies were less aware of the risks of pesticide use for the environment. A major issue is the persistence of pesticides in soils and waters. Indeed pesticides are biocides. Their lack of selectivity could lead to an important risk for living organisms and humans by contamination of drinking water and food. The presence of these biocides or their metabolites in soil, water, plants and even the atmosphere, together with their potential pharmacodynamic properties, can have harmful effects on the environment and on human health. In countries belonging to the European Union, regulations aim to reduce risks at the lowest level, but it is not the case everywhere. Some problems should now be overcome. Phytoremediation can reduce pollution and decrease the impact of pesticides on the environment. Two examples of substances are discussed in this review to illustrate the risk for the environment and remediation by plants to reduce it. First, the review focused on 1,1,1-trichloro-2,2,bis(p-chlorophenyl)ethane (DDT),an organochlorine insecticide used with a large success against human disease vectors or in crop protection against some coleopterans such as potato beetles. Its intensive use had contaminated huge areas in the world. Now, it is classified as a persistent organic pollutant (POP), due to its too slow degradation. Plants and associated microorganisms can degrade DDT but metabolites, dichlorodiphenyldichloroethylen (DDE), and dichlorodiphenyldichloroethan (DDD) are of identical persistence. The uptake by plants is very weak, and plant use could not resolve the DDT pollution. The second example is atrazine, an herbicide of the s-triazine group. It was largely used in crops such as maize. Now, atrazine and some metabolites are mainly pollutants of hydraulic networks. It is suspected to be an endocrine disruptor. Plants can help to reduce atrazine pollution by accelerating its microbial degradation but some degradative compounds, deethylatrazine (DEA) or deisopropylatrazine (DIA), polluted also water. However, plants could be useful to reduce water pollution because they can reduce run-off of atrazine derivates. Both examples showed the direct action of plants on pesticides by their capacity to take up, accumulate or detoxify organic substances or by their indirect action by stimulation of soil microbial activity in the breakdown of organic compounds. The use of plants is then presented in the form of examples describing their capacity to prevent pesticide pollution and the use of buffer zones between fields and hydraulic networks. The efficiency of vegetative filter strips (VFS) to protect water from pesticide run-off contamination leads the authorities to require them in good farming practice. Plants could be also used in the depuration of farming wastes. Macrophyte-planted constructed wetlands are efficient to purify farming wastes but their setting is critical. The variety of contaminated biotopes, as the number of pesticides to depurate, is large. This means that the plant choice must be done among many plants. High variability of plant tolerance does make choice more difficult. Three types of plants are particularly useful: graminae in buffer zones, trees such as poplar or willow in riparian zones or in phytoremediation processes due to large evapotranspiration capacities, and aquatic plants for waste depuration processes. The difficulties to find a polyvalent wild plant, lead to search for new methods to select plants more efficiently. The new genetic engineering technologies are a few developed because they can prove possible to broaden the scope even more. The conclusion consists of a brief glimpse of benefits of the use of plants and their limits.