It is indeed considered best practice nowadays to search publication databases for literature rather than only individual journals or conferences, typically combined with some snowballing. Scientific fields are manifold, and it is very easy to miss relevant publications if all you do is look through a pre-defined list of journals (independently of whether they are listed in Scopus, WoS, or whatever). Essentially, for a literature review to add value it needs to be able to make a case for completeness, and it is very hard to claim that without a fairly clear and comprehensive search procedure (along with rather objective acceptance criteria for papers).
I am in the last year of my thesis and I have to publish my work in an indexed journal. How to do so?
I should also say that this is, to no small degree, also a failing of your supervisor. Your supervisor should have told you early in the process that your research methodology has a good chance of not surviving peer review.
However, that doesn't really help you now. If you have repeatedly received this comment there really are only two ways forward:
- Do as you have been told, i.e., search the standard publication database(s) of your field and extend your review with any new material that may pop up.
- Lower your standards, i.e., go for a lower-ranked journal and hope that the reviewers there will be more happy with your approach.
As a sidenote:
every time I send my article to the journals I receive either major changes or a rejection
In my field (Computer Science), "major changes" is qualitatively different from "reject". A request for Major Revision is typically the first step towards getting accepted - it actually happens to most manuscripts. You really should not back out once you have received a request for major revision - usually, you address what can be addressed and argue about changes that cannot be addressed. In the vast majority of cases such papers end up being accepted.