Generally, the Musaceae is referred to as the "banana family", so all species of all 3 genera are called bananas. Every so often you will see plants of the genus Ensete and Musella referred to only as banana relatives, or being banana-like, but in the research community they are generally all called bananas. However, Ensete species (especially Ensete ventricosum) are also commonly called enset, but this is usually only when referring to them as a food crop (the corm is eaten in Africa and sometimes you will see phrases containing "banana and enset" to distinguish them).
The naming conventions for the wild species are straightforward Linnean binomials (ex,
Musa basjoo. There are about 70 described species of Musa, 6 Ensete and 2 Musella. However there are new species and subspecies of already known species of Musa being published frequently.
Edible bananas have their own naming system due to a long and complex history of multiple and sometimes complex hybridization events coupled with mutation and human selection. The history of the edible bananas is more or less known from comparing them genetically with the wild species, however it is still very complex and there is really only one common trait between all of them and that is parthenocarpy. Some edible bananas are hybrids (either between different wild species or between different subspecies of the same wild species), but the only effect this has had is in terms of diversity, it is not inherent in them being edible that they are hybrids. There is actually quite a large number of non-hybrid derived edible bananas.
There are 2 wild species that have primarily contributed genetically to edible bananas,
Musa acuminata and
Musa balbisiana, both of which have many different varieties of their own. Some edible bananas are derived entirely from M. acuminata varieties, and others from hybrids with M. balbisiana. The naming system of edible bananas works by describing their genome and ploidy level along with what is called a subgroup designation. All cultivars in the same subgroup should theoretically be monophyletic (though in practice there remains some discrepancies). For example, the Cavendish subgroup banana cultivar 'Gran Nain' is a triploid derived purely from M. acuminata, so its full proper written name would be Musa (AAA-Cavendish) 'Gran Nain'. The 'Dwarf Cavendish' would thus be Musa (AAA-Cavendish) 'Dwarf Cavendish'. 'Pisang Mas', a Sucrier subgroup diploid derived from M. acuminata would be Musa (AA-Sucrier) 'Pisang Mas'. Though there is some debate as to whether or not subgroups are valid for diploids since they are very diverse and many are so different that there would be many which have only one cultivar per subgroup is subgroup designation were to be given, so for now they are mainly just used with the triploids. There are also tetraploid edible bananas. These types of names are only used in written documents, when talking people normally just use the cultivar name and may mention the subgroup if it is unclear.
As for your 'Dwarf Orinoco', it is the Bluggoe subgroup which are derived from both M. acuminata and M. balbisiana. It is also a triploid. The full proper written name is Musa (ABB-Bluggoe) 'Dwarf Orinoco'. These designations are useful, however the complete history of the edible bananas is not known because the events which made them edible happened many times over thousands of years across a very large geographic area.
The term plantain is actually a very specific term which is often used in a very unspecific manner. Plantains are a subgroup of bananas, genetically AAB, and the name is often used synonymously with any type of banana which is cooked, but this is incorrect. They can be characterized with some technical morphological traits.
The website below has lots of information on banana taxonomy.
Musaceae
This diagram is meant to show the basic relative relationships between different groups within the Musaceae. The blue ovals are sections of the genus Musa.
This diagram is meant to show the origins of edible bananas. It is incomplete and in reality the origins are much more complex, however it is still useful in showing the manner in which different groups may have arisen.
