With the parenthetical remarks, Orberg is also illustrating something called "gerundive attraction," which happens when a gerund takes a direct object. That is, when you'd expect to see "cupidus epistulam scribendi" you will probably see "cupidus epistulae scribendae" in Latin texts.
The meaning is the same, but for some reason the grammar goes through a few twists and turns when a gerund takes an object (as one of my Latin professors explained, this happens because the Romans were all deviants and "attracted" to everything
). While a gerund could theoretically be used (and it does show up from time to time), Latin prefers to use a gerundive. However, changing from gerund to gerundive requires a few changes.
Two things happen in "gerundive attraction":
1. The gerund takes the gender of the noun it modifies. Hence, epistulam (f) + scribendi = scribendae (f). (This makes sense because a gerundive is an adjective and needs to agree with its noun.)
2. The noun (object of the gerund) adopts the case of the gerund! So while we'd expect epistulam to stay in the accusative case as the object, it changes to the genitive case to match the genitive of "scribendi." Hence, epistulam (acc) + scribendi (gen) = epistulae (gen).
Thus you have two different ways of saying the same thing. Cupidus epistulam scribendi / cupidus epistulae scribendae. Desirous of writing a letter.
When just translating literally from English to Latin, we'd probably default to the gerund (as literally matches our mode of expression), but Latin doesn't like it that way.