Review papers are long papers gathering the results of various researches and have many conclusions. When I want to address an issue or a conclusion about something should I repeatedly cite a Review paper and use points in that paper or I'd better find important papers on that topic and cite them?
I want to add to the other answers that, it happens sometimes that researchers read the review papers and not the original papers that are referred to in the review papers; but when they want to cite the content in their papers, they cite the original papers which they have never read.
The point is that, you should cite exactly the source you read. If you are reading that content from a review paper, you should cite that review paper. If you are reading the original paper, cite that original paper. Look at this example:
In case you only have read the review paper, not the original one:
In the review paper we read that the method x is perfectly applied to examine that [...]
In case that you read the review paper and then read the original paper:
In the review paper we read that the method x is perfectly applied to examine that [...]. It is clearly indicated in the original paper  that the method x only discusses parts of the instances of the experience including; instance a, instance b and instance c .
In my opinion, review papers are not always so precise and it happens that some content is misunderstood or wrongly copied in quotation and rephrasing in the review papers; so, if you find some content interesting, go to the references of the review paper and try to find and read the original papers as well.
Both. You cite a) the overview papers when referring to the suggested area and the comparison between the different methods and then you b) cite the original paper (when a method was introduced) when you refer to that individual method.
To add to Alexandros' answer. Refer to the review paper for the over-arching conclusions drawn in that paper. If you need to refer to results emanating from an individual paper, you should refer to that particular paper. Remember that you should cite the source and you should avoid citing work through others. The purpose of a review paper varies but usually involve summarizing, synthesizing and finding new structure or gaps in knowledge from a large body of literature. These are the "conclusions" from such a paper which you should cite. It is also common to cite reviews as good sources for literature on a particular topic.
As with any research there is no guarantee all material used in a review paper is well represented, mistakes or misrepresentations occur. It is therefore necessary to go back to original works if you cite material that is key to your own paper.