How do I wait for 2 actions to dispatch another action - reactjs

Question

I'm looking for a way to dispatch an action once I know multiple actions have been dispatched. I know that I can do this now by storing some variables in store. But is there a better way of doing it (Saga, Discoverables) etc. I tried to go through their documentation, but couldn't understand if it can be used for this purpose.
E.g in the below code, I want to dispatch some action in middleware when I know that 2 required actions have dispatched
import React, { Component } from 'react';
import ReactDOM from 'react-dom'
import {createStore, applyMiddleware} from 'redux'
import {Provider} from 'react-redux'
const dummyReducer = (state=null,action) => {
return state
}
const defaultAction1 = { type: 'DEFAULT_ACTION1'}
const requiredAction1 = { type: 'REQUIRED_ACTION1'}
const requiredAction2 = { type: 'REQUIRED_ACTION2'}
const middle = (store) => (next) => (action) => {
if(action.type === 'REQUIRED_ACTION1')
console.log('Required actions satisfied')
next(action)
}
const store = createStore(dummyReducer, applyMiddleware(middle))
store.dispatch(defaultAction1)
//Dispatch an action after 2 secs
setTimeout(store.dispatch.bind(null,requiredAction1),2000)
setTimeout(store.dispatch.bind(null,requiredAction2),4000)
class DummyComponent extends Component{
render(){
return(<h1> nothing to see here </h1>)
}
}
ReactDOM.render(<Provider store={store}>
<DummyComponent />
</Provider>,
document.getElementById('root')
);

Answer 1

You can do this in saga simply by:
yield all([
take(REQUIRED_ACTION1),
take(REQUIRED_ACTION2),
])
yield put({type: FINAL_ACTION})

Answer 2

Actions must be self-explanatory and complete, it's a good practice to dispatch actions with all the needed info in their payloads.
Another reason that you should always try is dispatching different actions based on their purposes and take advantage of Redux tooling to debug your application properly. Redux updates the states synchronously and this is actually an advantage of its design. That said you know when one action goes through all the reducers you will have the proper state to reflect the changes on your UI part.
If you rely on two actions you somewhat can't rely on a single source of the truth which is your action1 and action2. And this forces you to wait for both of these in a middleware and it turns your actions into effects and effects are used in a different context.