In future, do-it-yourself (DIY) by open knowledge access and repositories will gain more publishing momentum (e.g. Elsevier has bought into SSRN). We will see many new intersections of 'publishing rituals'.
In my point of opinion, Journals indexed either in SCOPUS or WEB OF SCIENCE are good even impact factor below one. While one more thing if the journal is open access then it reaches en mass.
This may depend on your discipline. You must also understand that there are some bad journals as under some of these so called big publishing houses but I will recommend Nature and AIP for science disciplines (particularly Materials science and Physics)