How should I use Sequential and Compose classes of albumentations package?

Question

I want to combine some transformations and then concatenate them. It's comfortable for me to create a composition of some transformations and then combine these compositions into a one transformation. I'm expecting, that when I'll apply this transformation to dataset, all basic transformations will be applied sequentially in order, in which them are in code.

Example:

transform_resize = A.Sequential([
    A.Resize(200),
])

some_transforms = A.Sequential([
    transform1,
    transform2,
    transform3
])

transform_norm = A.Sequential([
    A.Normalize(mean=mean, std=std),
    ToTensorV2()
])

transform = A.Compose([
    transform_resize,
    some_transforms,
    transform_norm
])

Expected sequence of transformations, when transform will be applied to dataset:

[Resize, transform1, transform2, transform3, Normalize, ToTensorV2]

The questions are:

1. Is my way to use Compose and Sequential correct?
2. What are the rules to use these two classes inside each other? Sequential class documentation only said:

This transform is not intended to be a replacement for Compose. Instead, it should be used inside Compose the same way OneOf or OneOrOther are used

but it's nothing about using Sequential inside Sequential and Compose inside Compose