Yes that is helpful, thank you!
Because it's not limited to "gods" I think it's worth spelling out the reasoning issue. As I see it:
1. Cotta is alleging that Epicurus said that in order to understand something, you have to have seen a prior example and understood the prior example. Implicitly the point is "you have to have examined it here on earth in your own experience and understood it before you can apply the generalization to a new example."
2. Cotta alleges that there is nothing like the sun, moon, or stars here on earth, so Epicurus has had no examples to examine.
3. Cotta alleges that since Epicurus had no examples, and under Epicurus's rules that we require examples on which to reason, we cannot make any reliable statements about the sun moon or stars. But Cotta knows that Epicurus DOES claim to know that the sun and stars and planets are not gods.
4. Cotta says this position that the suns and stars and planets are not gods is contradictory. Since Epicurus has had no examples of suns or moons or planet to examine, Epicurus (allegedly under his own rules) should not be making any representations about the sun and stars and planets at all. Cotta thinks he has Epicurus in a trap, so he says "Epicurus if you are willing without evidence reach the conclusion that sun stars and planets are not gods, why don't you go ahead and admit without evidence the sun stars and planets are gods, and that gods can exist in other than human form?"
So I see Cotta as trying to take advantage of the argument that we also see in Lucretius, where Lucretius says "the gods could not have made the universe because they had no pattern to go by." It was apparently understood that the Epicureans argued regularly that you must have a pattern from experience in order to understand something and work with it.
Any skeptic who thinks that Epicureans demand examples before they believe in something will attack any Epicurean position on the sun moon and stars by saying : "Under your own theory of reasoning, you can't say anything at all about something unless you have seen a prior example of it, so you shouldn't be talking at all about the sun moon or stars."
And that sounds like a reasonable argument against someone who is alleging you can only reason based on analogy from sensory experience. But it's not a persuasive argument against Epicurus, because the Epicurus' position is not that you reason based ONLY on past experience, your reason can also be based on reasonable inferences from circumstantial evidence. In that way it is permissible to make conclusions about things you haven't seen, because you can infer new possibilities that you haven't seen based on other examples of what we have seen.
This is how the difference plays out in court:
If a judge in court only allows "direct evidence," then he will allow into evidence a witness who says "I saw Tom Jones strangle Sally Smith in the jail cell."
If a court allows only direct evidence, the judge will EXCLUDE the testimony of a witness who says: "I saw Tom Jones and Sally Smith alone in the jail cell, and then I left, and ten minutes later I came back and Sally Smith was dead with red marks around her neck, and no one had disturbed the lock on the cell."
The latter testimony is not "direct" evidence" but "circumstantial evidence." Circumstantial evidence is controversial and has to be treated carefully. But our legal system has decided that it is reasonable to ally juries to consider circumstantial evidence in court, because sometimes direct evidence is not available but circumstantial evidence is very strong.
The answer to Cotta is that Epicurean philosophy doesn't require direct evidence all the time, such as in the case of the existence of atoms or what happens to self-consciousness after death. Epicurean philosophy allows inference based on circumstantial evidence, and it considers the conclusions of persuasive amounts of circumstantial evidence to be as worthy of reliance, even in life-and-death decisionmaking, as evidence from that which has been observed directly.
Therefore I see it as essential to point out that Cotta's is wrong to allege that Epicurus requires direct evidence to support decisionmaking. After that, we also need to take the next step of discussing the proper use of circumstantial evidence, so we can explain how he's misrepresenting Epicurus, because that's not obvious to everyone.
And it's also worth pointing out another inconsistency in Cotta:
Why, Cotta, if you are a skeptic and don't think anything is knowable, do you agree with Velleius that gods are happy, and gods have virtue, and gods use reason? Where do you come by that knowledge?
In fact we need to ask that question of both Cotta and Epicurus in order to understand the big picture of how their reasoning differs. If we just stop and say "this topic is about gods and I don't care about gods" then we will cut ourselves off from major aspects of how Epicurean reasoning works on everything, not just gods.