Perhaps it will be helpful to look at examples where his advice is explicitly ignored. There are a number of ways in which our common use of language intentionally relies NOT on the "first mental image", but on some other aspect or quality for aesthetic, poetic, or rhetorical effect. For example;
Synechdoche: a figure of speech by which a part is put for the whole (such as fifty sail for fifty ships), the whole for a part (such as society for high society), the species for the genus (such as cutthroat for assassin), the genus for the species (such as a creature for a man), or the name of the material for the thing made (such as boards for stage)
Metonymy: a figure of speech consisting of the use of the name of one thing for that of another of which it is an attribute or with which it is associated (such as "crown" in "lands belonging to the crown")
Metaphor: a figure of speech in which a word or phrase literally denoting one kind of object or idea is used in place of another to suggest a likeness or analogy between them (as in drowning in money)
Epicurus seems to greatly dislike these devices; certainly with regards to Philosophy, and possibly in general. Lucretius, as a masterful poet, is guilty of using all of them! In the very second line of the poem, for example, he refers to the stars or constellations as the "sliding signs of heaven". He's a materialist; he doesn't actually believe that the random clusters of stars are meaningful signs or representations. But the line reads beautifully, and the phrase serves the meter of the poem, so he uses it.
Here's a good example of why this can be a huge problem: in the Gospel of Matthew, 19:24, it is said, "it is easier for a camel to go through the eye of a needle than for a rich man to enter the kingdom of God."
Well, it turns out that Christians have a strong material interest in denying the plain reading of everything Jesus ever said about wealth. This passage is no different; and so they have invented out of wholecloth a theory that the "Eye of the Needle" is a figure of speech for a pedestrian portal or doorway through a wall, adjacent to the main large gate. This usage is completely unattested in classical literature. It might turn out to be correct, but we have absolutely no way of knowing. If Jesus wished to be understood, he might have taken Epicurus' advice!
Another problem relates to how we define words—how we describe language using language. In one amusing story, Diogenes the Cynic elbows his way into the Academy with a plucked chicken under his arm. Aristotle had defined "Man" as a "featherless biped"; Diogenes lifted the chicken, and proclaimed his discovery of Aristotle's Man. As an Epicurean, I think there were two errors on display here; one was to define a word so broadly so as to be meaningless (Aristotle's mistake). The other was to mock the original effort without furnishing a constructive alternative, and to poison the well for everyone with ridicule (Diogenes' mistake).