I have written an article in Persian and then translated it into English. Is it possible to send them to two different journals (one in Iran and the other one in Belgium)?
Yes, this is a problem. You should not submit the same work to more than one journal at a time. This applies regardless of language - merely being in a different language does not make the actual work unique.
Further, consider some practical considerations - a simultaneous submission means that you'll have to go through peer-review for both papers and (assuming you don't get caught and summarily rejected) presumably have to make some changes. What if the changes don't align? Now you don't have two translations of the same paper, but two subtly different papers in two languages.
If you're genuinely interested in making your paper more accessible, either post a translation online, or wait until one paper is accepted and then approach a journal in the other language about posting a translation.
If you write an article in Persian, you have the legal right to translate it to English, as long as you haven't assigned the copyright to someone else (in which case, you have to obtain the right to create a derivative work from the copyright owner). Many journals require you to assign copyright when they publish an article, and require you to avow that you (still) have the right to assign that copyright. So if you have assigned the right to make copies (including translations) to one journal, you can't then truthfully vow to the other journal that you can assign the right to the derivative work in English.
Journals have two basic reasons for following a no double publishing rule. One is the aforementioned copyright problem, that it might put one journal in a bind with respect to copyright. If neither journal requires copyright assignment and only requires a non-exclusive license, then the second reason may also be applicable, which is that journals consider their space to be a valuable commodity, to be used only when there is sufficient justification. Basic academic quality is one reason, but novelty is another -- a journal's editor could well decide that it is not worthwhile to use journal space to publish a paper that has already been published.
Since this is a matter of editorial judgment, the solution is simply to ask each editor whether it is okay for you to also submit a version in the other language. If both editors agree (and assuming that you have stated the facts clearly enough), then there would be nothing problematic with submitting a version of a paper in Farsi to one journal and a version in English to another journal. Versions in French and English would be more problematic, because there is a reasonable expectation that scholars can read both languages, whereas there are relatively few areas where scholars are standardly expected to read both English and Farsi, and providing an English translation could be a significant contribution.
So, to repeat, ask the editors.
Publishing the same manuscript in more than one journal at a time is technically double submission. But this wouldn't be so if you consider one as a translated version of the other after one is published as long as you make this explicitly clear in the second paper. It isn't ethically wrong to publish your research in different language as it increases the visibility of your research in the right sense.
I do not recommend it, for the following reasons:
Firstly, because the same ideas and results will appear on two different papers, and someone that wants to credit your findings (and reads both languages) doesn't know which paper to cite.
Secondly, if I would see your Iranian paper (in journal X) cited on paper Y but I would only know the English version of it (in journal Z), I would have to check that they are, in fact, the same scientific result.
Finally, in most journals publications are required to present new results. Pushing your case to the edge, I could write 2 papers in English: one in such a way that only a native speaker would understand, and another that my community would understand. This is not acceptable by most journals.
There are exceptions on which publishing the same results is acceptable, for instance, some journals do not require the results to be new. In this situation, I don't see the issue.