In her 1993 Preface to The Golden Notebook, Nobel laureate Doris Lessing writes:
"Currently I am writing volume one of my autobiography, and ... I have to conclude that fiction is better at 'the truth' than a factual record. Why this should be so is a very large subject and one I don't begin to understand." (p. ix)
I cannot say that I understand it either. However, I wanted to explore the notion that it might be related to idealization and abstraction, which have interested philosophers of science over the last few decades.
Roughly, an idealization is something that is strictly speaking false. For example, in usual derivations of the ideal gas law (PV=nRT) from statistical mechanics, the gas molecules are assumed to be massless point particles. But everyone agrees that, actually, the gas molecules do have some mass and some volume. But this reveals something: the mass or volume of the constituent particles do not play a (significant) role in the gas's macro-level properties (like pressure and temperature).
Roughly, an abstraction leaves out something (as opposed to putting in something false); for example, when doing classical mechanics, we need to know the mass and velocity of bodies, but we do not need to know their colors. And in many cases, we don't need to factor in their shapes. This is like the idealization case, because this reveals that the shape of a body is irrelevant to its behavior after an elastic collision.
Now, fiction is also full of both straightforward falsehoods (there's no such person as Harry Potter) and abstractions (How many hairs, exactly, are on Harry Potter's head?). Perhaps we can 'begin to understand' the fact that Lessing finds so perplexing by saying that these two elements of fiction play a role similar to idealization and abstraction in science. That is, by stripping away the infinitely many traits that concrete, actual things and events have, fiction makes manifest the explanatory (perhaps causal?) core of the people and circumstances it depicts, leaving out (supposed) irrelevancies. So I would not say (as Lessing does "that fiction is better at 'the truth' than the factual record," which sounds paradoxical, but rather that fiction can provide a better explanation than a detailed diary -- and it does so for the same reason that idealization and abstraction in science provide better explanations than would a complete causal history of the universe.