research at early time started focusing on 5s, however when 16s was sequenced it showed more reliable and competitive results, size more than 5s is advantageous. even further improvement in sequencing techniques lead to wide use of 16S. however 23S is also an option, researchers have started using it and papers are available but size is more big and take more time to sequence full gene. still 16S is good choice and reliable.
As others have stated, the 5s doesn't carry as much information due to it's short sequence and lack of both conserved and variable regions that the 16s contains. The 23s may be an option but probably doesn't have some the advantages that the 16s gene has. The beauty of the 16s gene is that the distance between conserved and variable regions makes it applicable to the various sequence technologies available, which often only sequence short 200-600bp fragments of the gene, which may not be the case for the 23s gene. The 16s gene also has regions that are conserved with the 18s allowing a wider variety of organisms to be captured with the same "universal primer" sets so that yeasts and other eukaryotes can also be surveyed alongside the bacteria and archaea from the same sample. And finally, regardless of the historical reasons as to why 16s was originally chosen, it is now the ideal gene of choice because it has been so well characterized and there are a plethora of curated databases with the phylogenetic information to assess taxonomy within a sample. The 16s isn't perfect though, some species can't be differentiated by 16s because the gene is conserved in both species and it's important to choose the correct variable region of the 16s gene to distinguish between certain species of interest.